U.S. patent application number 17/040106 was filed with the patent office on 2021-01-28 for mems device manufacturing method, mems device, and shutter apparatus using the same.
The applicant listed for this patent is SUMITOMO PRECISION PRODUCTS CO., LTD.. Invention is credited to Mario KIUCHI, Gen MATSUOKA.
Application Number | 20210024352 17/040106 |
Document ID | / |
Family ID | 1000005193307 |
Filed Date | 2021-01-28 |
![](/patent/app/20210024352/US20210024352A1-20210128-D00000.png)
![](/patent/app/20210024352/US20210024352A1-20210128-D00001.png)
![](/patent/app/20210024352/US20210024352A1-20210128-D00002.png)
![](/patent/app/20210024352/US20210024352A1-20210128-D00003.png)
![](/patent/app/20210024352/US20210024352A1-20210128-D00004.png)
![](/patent/app/20210024352/US20210024352A1-20210128-D00005.png)
![](/patent/app/20210024352/US20210024352A1-20210128-D00006.png)
United States Patent
Application |
20210024352 |
Kind Code |
A1 |
MATSUOKA; Gen ; et
al. |
January 28, 2021 |
MEMS DEVICE MANUFACTURING METHOD, MEMS DEVICE, AND SHUTTER
APPARATUS USING THE SAME
Abstract
Provided is a method including at least the thermal treatment
step of thermally treating a SOI substrate having a first silicon
layer at a first temperature that the diffusion flow rate of an
interstitial silicon atom in a silicon single crystal is higher
than the diffusion flow rate of an interstitial oxygen atom and the
processing step of processing the SOI substrate after the thermal
treatment step to obtain a displacement enlarging mechanism.
Inventors: |
MATSUOKA; Gen; (Hyogo,
JP) ; KIUCHI; Mario; (Hyogo, JP) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
SUMITOMO PRECISION PRODUCTS CO., LTD. |
Hyogo |
|
JP |
|
|
Family ID: |
1000005193307 |
Appl. No.: |
17/040106 |
Filed: |
February 22, 2019 |
PCT Filed: |
February 22, 2019 |
PCT NO: |
PCT/JP2019/006735 |
371 Date: |
September 22, 2020 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
B81B 2201/042 20130101;
B81B 3/0072 20130101; B81C 2201/017 20130101; B81B 2201/031
20130101; G02B 26/04 20130101; B81C 1/00666 20130101 |
International
Class: |
B81C 1/00 20060101
B81C001/00; B81B 3/00 20060101 B81B003/00; G02B 26/04 20060101
G02B026/04 |
Foreign Application Data
Date |
Code |
Application Number |
Mar 28, 2018 |
JP |
2018-061103 |
Claims
1. A MEMS device manufacturing method comprising: at least a
thermal treatment step of thermally treating a substrate having a
silicon layer at a first temperature that a diffusion flow rate of
an interstitial silicon atom in a silicon single crystal is higher
than a diffusion flow rate of an interstitial oxygen atom; and a
processing step of processing the substrate after the thermal
treatment step to obtain a MEMS device.
2. The MEMS device manufacturing method according to claim 1,
wherein at the step performed after the thermal treatment step, a
temperature applied to the silicon layer is equal to or lower than
a second temperature that the diffusion flow rate of the
interstitial oxygen atom in the silicon single crystal is higher
than the diffusion flow rate of the interstitial silicon atom and
precipitated oxide contained in the silicon layer does not
substantially grow.
3. The MEMS device manufacturing method according to claim 1,
wherein the substrate is a multilayer bonded substrate configured
such that a handle layer, an insulating layer, and a device layer
are stacked on each other in this order, and the device layer is
the silicon layer, and is formed using a silicon substrate
manufactured by a Czochralski (CZ) method.
4. The MEMS device manufacturing method according to claim 1,
wherein the silicon layer contains a predetermined concentration of
oxygen, and the predetermined concentration is in a range of
5.times.10.sup.17/cm.sup.3 to 1.times.10.sup.18/cm.sup.3.
5. The MEMS device manufacturing method according to claim 1,
further comprising: a substrate preparation step of preparing a
substrate having a silicon layer whose stacking fault density is
equal to or greater than 1.times.10.sup.4/cm.sup.2, wherein at the
processing step, the substrate prepared at the substrate
preparation step is processed to obtain the MEMS device.
6. The MEMS device manufacturing method according to claim 1,
further comprising: a substrate preparation step of preparing a
substrate having a silicon layer whose precipitated oxide density
is equal to or less than 5.times.10.sup.5/cm.sup.2, wherein at the
processing step, the substrate prepared at the substrate
preparation step is processed to obtain the MEMS device.
7. The MEMS device manufacturing method according to claim 1,
further comprising: a substrate preparation step of preparing a
substrate having a silicon layer whose precipitated oxide density
is equal to or less than 5.times.10.sup.5/cm.sup.2 and whose
stacking fault density is equal to or greater than
1.times.10.sup.4/cm.sup.2, wherein at the processing step, the
substrate prepared at the substrate preparation step is processed
to obtain the MEMS device.
8. The MEMS device manufacturing method according to claim 5,
wherein at the step performed after the substrate preparation step,
a temperature applied to the silicon layer is equal to or lower
than a second temperature that a diffusion flow rate of an
interstitial oxygen atom in a silicon single crystal is higher than
a diffusion flow rate of an interstitial silicon atom and
precipitated oxide contained in the silicon layer does not
substantially grow.
9. The MEMS device manufacturing method according to claim 5,
wherein the substrate is a multilayer bonded substrate configured
such that a handle layer, an insulating layer, and a device layer
are stacked on each other in this order, the device layer is the
silicon layer, and is formed using a silicon substrate manufactured
by a Czochralski (CZ) method, and at the step performed after the
substrate preparation step, the temperature applied to the silicon
layer is equal to or lower than the second temperature that the
diffusion flow rate of the interstitial oxygen atom in the silicon
single crystal is higher than the diffusion flow rate of the
interstitial silicon atom and the precipitated oxide contained in
the silicon layer does not substantially grow.
10. The MEMS device manufacturing method according to claim 5,
wherein the silicon layer contains a predetermined concentration of
oxygen, and the predetermined concentration is in a range of
5.times.10.sup.17/cm.sup.3 to 1.times.10.sup.18/cm.sup.3.
11. The MEMS device manufacturing method according to claim 5,
wherein the processing step includes at least a mask pattern
formation step of forming a mask pattern for processing the silicon
layer; and a silicon layer processing step of patterning the
silicon layer by means of the mask pattern, and the mask pattern
includes a thermally-oxidized film formed on a surface of the
substrate at the thermal treatment step.
12. The MEMS device manufacturing method according to claim 1,
wherein the MEMS device includes at least a thermal actuator
configured to generate heat by current application to displace in a
predetermined direction according to a generated heat temperature
and a drive target member coupled to the thermal actuator.
13. A MEMS device comprising: at least a substrate having a silicon
layer; a fixing portion formed on the substrate; a thermal actuator
coupled to the fixing portion and configured to generate heat by
current application to displace in a predetermined direction
according to a generated heat temperature; and a drive target
member coupled to the thermal actuator, wherein a precipitated
oxide density of the silicon layer is equal to or less than
5.times.10.sup.5/cm.sup.2, and a member forming the thermal
actuator is the silicon layer.
14. A MEMS device comprising: at least a substrate having a silicon
layer; a fixing portion formed on the substrate; a thermal actuator
coupled to the fixing portion and configured to generate heat by
current application to displace in a predetermined direction
according to a generated heat temperature; and a drive target
member coupled to the thermal actuator, wherein a stacking fault
density of the silicon layer is equal to or greater than
1.times.10.sup.4/cm.sup.2, and a member forming the thermal
actuator is the silicon layer.
15. A MEMS device comprising: at least a substrate having a silicon
layer; a fixing portion formed on the substrate; a thermal actuator
coupled to the fixing portion and configured to generate heat by
current application to displace in a predetermined direction
according to a generated heat temperature; and a drive target
member coupled to the thermal actuator, wherein a precipitated
oxide density of the silicon layer is equal to or less than
5.times.10.sup.5/cm.sup.2, and a stacking fault density of the
silicon layer is equal to or greater than
1.times.10.sup.4/cm.sup.2, and a member forming the thermal
actuator is the silicon layer.
16. The MEMS device according to claim 13, wherein the substrate is
a multilayer bonded substrate configured such that a handle layer,
an insulating layer, and a device layer are stacked on each other
in this order, and the device layer is the silicon layer, and is
formed using a silicon substrate manufactured by a Czochralski (CZ)
method.
17. The MEMS device according to claim 13, wherein the silicon
layer contains a predetermined concentration of oxygen, and the
predetermined concentration is in a range of
5.times.10.sup.17/cm.sup.3 to 1.times.10.sup.18/cm.sup.3.
18. The MEMS device according to claim 13, wherein a member forming
the drive target member includes the silicon layer.
19. A shutter apparatus for closing or opening an optical path by a
drive target member, comprising: the MEMS device according to claim
13; and first and second electrodes arranged on the fixing portion
and electrically connected to both end portions of the thermal
actuator.
Description
TECHNICAL FIELD
[0001] The present invention relates to a MEMS device manufacturing
method, a MEMS device, and a shutter apparatus using the MEMS
device.
BACKGROUND ART
[0002] Typically, an optical apparatus configured to change an
optical path by means of a MEMS (micro electro mechanical systems)
device has been known. Patent Document 1 discloses a technique in
which an optical waveguide coupled to a thermal actuator and
provided movable in a direction parallel to a substrate surface is
arranged between other optical waveguides fixed apart from each
other on the substrate surface and an optical path between the
optical waveguides arranged in a fixed manner is switched by drive
of the thermal actuator. Note that the above-described MEMS device
is manufactured using a SOI (silicon on insulator) substrate having
a silicon oxide film layer between two silicon single-crystal
layers.
CITATION LIST
Patent Document
[0003] PATENT DOCUMENT 1: U.S. Pat. No. 7,298,954
SUMMARY OF THE INVENTION
Technical Problem
[0004] Generally, in terms of a manufacturing cost etc., a
so-called bonded SOI substrate formed in such a manner that another
silicon substrate is bonded to an oxide layer formed on a surface
of a silicon substrate has been often used as a SOI substrate used
for manufacturing a MEMS device. Similarly, in terms of the
manufacturing cost etc., these silicon substrates are manufactured
using a Czochralski (CZ) method.
[0005] However, when the MEMS device as described in Patent
Document 1 is produced using such a SOI substrate as a starting
base material, precipitated oxide is formed at high density in a
silicon layer depending on the concentration of oxygen contained in
the silicon layer and thermal treatment conditions in SOI substrate
production or MEMS device manufacturing, and dislocation due to
such precipitated oxide might occur. Such dislocation occurs at
high density in the silicon layer, leading to a probability that
plastic deformation is caused in a case where the thermal actuator
has been used for a long period of time. Moreover, due to such
plastic deformation, there is a probability that operation
reliability of the MEMS device and therefore operation reliability
of a shutter apparatus are degraded.
[0006] The present invention has been made in view of such a point,
and an object thereof is to provide a highly-reliable MEMS device
configured such that formation of dislocation due to precipitated
oxide in a silicon layer is suppressed and a shutter apparatus
including the MEMS device.
Solution to the Problem
[0007] For accomplishing the above-described object, a MEMS device
manufacturing method according to the present invention includes at
least the thermal treatment step of thermally treating a substrate
having a silicon layer at a first temperature that the diffusion
flow rate of an interstitial silicon atom in a silicon single
crystal is higher than the diffusion flow rate of an interstitial
oxygen atom and the processing step of processing the substrate
after the thermal treatment step to obtain a MEMS device.
[0008] According to this method, generation of precipitated oxide
in the silicon layer can be suppressed, and the density thereof can
be decreased. With this configuration, occurrence of dislocation
due to the precipitated oxide can be suppressed, and operation
reliability of a MEMS device can be enhanced.
[0009] Moreover, a MEMS device according to the present invention
includes at least a substrate having a silicon layer, a fixing
portion formed on the substrate, a thermal actuator coupled to the
fixing portion and configured to generate heat by current
application to displace in a predetermined direction according to a
generated heat temperature, and a drive target member coupled to
the thermal actuator, the precipitated oxide density of the silicon
layer being equal to or less than 5.times.10.sup.5/cm.sup.2 and a
member forming the thermal actuator being the silicon layer.
[0010] According to this configuration, plastic deformation in
long-term use of the thermal actuator including the silicon layer
as a component can be suppressed, and the operation reliability of
the MEMS device can be enhanced.
[0011] A shutter apparatus according to the present invention
closes or opens an optical path by a drive target member, and
includes the above-described MEMS device and first and second
electrodes arranged on the fixing portion and electrically
connected to both end portions of the thermal actuator.
[0012] According to this configuration, the operation reliability
of the MEMS device can be enhanced, and therefore, operation
reliability of the shutter apparatus itself having the MEMS device
can be enhanced.
Advantages of the Invention
[0013] As described above, according to the present invention,
occurrence of dislocation in the silicon layer can be suppressed,
and the operation reliability of the MEMS device can be
enhanced.
BRIEF DESCRIPTION OF THE DRAWINGS
[0014] FIG. 1 is a plan view of a shutter apparatus according to a
first embodiment.
[0015] FIG. 2 is a sectional view of the shutter apparatus along an
II-II line in FIG. 1.
[0016] FIG. 3 is a plan view of the shutter apparatus in a drive
state.
[0017] FIG. 4A is a sectional view illustrating one step in the
method for manufacturing a SOI substrate.
[0018] FIG. 4B is a sectional view illustrating one step in the
method for manufacturing the SOI substrate.
[0019] FIG. 4C is a sectional view illustrating one step in the
method for manufacturing the SOI substrate.
[0020] FIG. 5A is a sectional view illustrating one step in the
method for manufacturing the shutter apparatus.
[0021] FIG. 5B is a sectional view illustrating one step in the
method for manufacturing the shutter apparatus.
[0022] FIG. 5C is a sectional view illustrating one step in the
method for manufacturing the shutter apparatus.
[0023] FIG. 5D is a sectional view illustrating one step in the
method for manufacturing the shutter apparatus.
[0024] FIG. 6 is a table showing a precipitated oxide density, a
stacking fault density, and HTOL test results in the case of
different thermal treatment temperatures at a thermal oxidization
step.
DESCRIPTION OF EMBODIMENTS
[0025] Hereinafter, embodiments of the present invention will be
described in detail with reference to the drawings. Description of
the preferable embodiments below will be set forth merely as
examples in nature, and is not intended to limit the present
invention, an application thereof, and a use application thereof at
all.
Embodiment
[0026] [Configuration of Shutter Apparatus]
[0027] FIG. 1 illustrates a plan view of a shutter apparatus 1
according to one embodiment of the present invention, and FIG. 2
illustrates a sectional view of the shutter apparatus 1 along an
II-II line in FIG. 1. Note that, e.g., the dimensions, thickness,
and detailed shape of each member drawn in the figure are, in some
cases, different from actual dimensions, thickness, and shape.
[0028] The shutter apparatus 1 includes a fixing portion 2, a first
actuator 3 and a second actuator 4 coupled to the fixing portion 2,
a first beam 5 having a first end portion 5a and a second end
portion 5b and configured such that the first end portion 5a is
coupled to the first actuator 3, a second beam 6 having a third end
portion 6b and a fourth end portion 6c and configured such that the
third end portion 6b is coupled to the second actuator 4, a drive
target member 7 coupled to the second end portion 5b of the first
beam 5 and the fourth end portion 6c of the second beam 6, a first
electrode 101, and a second electrode 102.
[0029] Hereinafter, a longitudinal direction of the first beam 5
will be, for the sake of convenience in description, referred to as
a "X-direction," a longitudinal direction of the first actuator 3
and the second actuator 4 will be referred to as a "Y-direction,"
and a thickness direction of the shutter apparatus 1 will be
referred to as a "Z-direction." Note that in the X-direction, a
left side in FIG. 1 will be sometimes merely referred to as a "left
side," and a right side in FIG. 1 will be sometimes merely referred
to as a "right side." In the Y-direction, an upper side in FIG. 1
will be sometimes merely referred to as an "upper side," and a
lower side in FIG. 1 will be sometimes merely referred to as a
"lower side." In the Z-direction, an upper side in FIG. 2 will be
sometimes referred to as an "upper surface," and a lower side in
FIG. 2 will be sometimes referred to as a "lower surface."
Moreover, the first end portion 5a of the first beam 5 or the third
end portion 6b of the second beam 6 will be sometimes referred to
as a "base end," and the second end portion 5b of the first beam 5
or the fourth end portion 6c of the second beam 6 will be sometimes
referred to as a "tip end."
[0030] The shutter apparatus 1 is a so-called MEMS shutter, and is
manufactured by a micromachining technique to which a semiconductor
micromachining technique is applied. The shutter apparatus 1 is
manufactured using a SOI substrate 200. The SOI substrate 200 is
configured such that a first silicon layer 210 made of
single-crystal silicon, an oxide film layer 220 made of SiO.sub.2,
and a second silicon layer 230 made of single-crystal silicon are
stacked on each other in this order.
[0031] As described above, in the shutter apparatus 1, the fixing
portion 2, the first actuator 3, the second actuator 4, the first
beam 5, the second beam 6, and the drive target member 7 are
integrally molded with a silicon material, thereby forming a
displacement enlarging mechanism 10 (a MEMS device). Note that the
oxide film layer 220 and the second silicon layer 230 remain only
on the lower surfaces of a first base member 21 and a second base
member 22 forming the fixing portion 2, and the oxide film layer
220 and the second silicon layer 230 at the lower surfaces of the
first actuator 3, the second actuator 4, the first beam 5, the
second beam 6, and the drive target member 7 as movable members are
removed at a later-described manufacturing process.
[0032] As illustrated in FIG. 1, the shutter apparatus 1 has, for
example, a rectangular overall shape as viewed in plane. The fixing
portion 2 is a frame forming such a rectangular overall shape of
the shutter apparatus 1 as viewed in plane. The fixing portion 2
includes the first base member 21 and the second base member 22
arranged facing the Y-direction.
[0033] Note that the fixing portion 2 is divided into two parts of
the first base member 21 and the second base member 22 in the first
silicon layer 210, but is connected as one in the oxide film layer
220 and the second silicon layer 230. Thus, the relative positions
of the first base member 21 and the second base member 22 are
fixed, and the first base member 21 and the second base member 22
can support the movable members.
[0034] Further, a first recessed portion 21a of the first base
member 21 and a third recessed portion 22a of the second base
member 22 are arranged at such positions that openings thereof face
each other, and form a substantially-rectangular opening 20L
elongated in the Y-direction and provided for arrangement of the
first actuator 3. Similarly, a second recessed portion 21b of the
first base member 21 and a fourth recessed portion 22b of the
second base member 22 are arranged at such positions that openings
thereof face each other, and form a substantially-rectangular
opening 20R elongated in the Y-direction and provided for
arrangement of the second actuator 4.
[0035] As described above, the fixing portion 2 is in a shape
covering a widest possible area while movable areas of the movable
members can be ensured, and therefore, ensures high stiffness
necessary as the frame supporting the first actuator 3 and the
second actuator 4.
[0036] The first actuator 3 includes two actuators 31 arranged in
parallel. Two actuators 31 are rod-shaped thermal actuators
extending in the Y-direction. As described later, the actuators 31
are formed in such a manner that the first silicon layer 210 is
patterned and the oxide film layer 220 and the second silicon layer
230 positioned below the first silicon layer 210 are removed. The
temperatures of the actuators 31 themselves increase due to Joule
heat generated by current application to the actuators 31
themselves, and according to such a temperature increase, the
actuators 31 curve in a predetermined direction. Moreover, the
actuators 31 are coupled to each other at an intermediate portion
3a at the substantially center between a first end portion 3b and a
second end portion 3c of the first actuator 3 in the longitudinal
direction. As described above, two actuators 31 are coupled to each
other at the intermediate portion 3a, and therefore, drive force of
two actuators 31 is bound so that the first actuator 3 can exert
great drive force. As described later, the first actuator 3 is
thermally expanded by heating by power distribution to generate the
drive force.
[0037] The first end portions 3b of two actuators 31 are coupled to
the first base member 21 at a bottom portion of the first recessed
portion 21a of the first base member 21. The second end portions 3c
of two actuators 31 are coupled to the second base member 22 at a
bottom portion of the third recessed portion 22a of the second base
member 22.
[0038] In a precise sense, the first actuator 3 does not extend
straight in the Y-direction, but slightly bends such that the
intermediate portion 3a protrudes leftward in the X-direction as a
drive direction of the first actuator 3 or slightly curves to
entirely expand leftward in the X-direction.
[0039] The second actuator 4 includes two actuators 41 arranged in
parallel. Two actuators 41 are rod-shaped members extending in the
Y-direction, and are thermal actuators similar to the first
actuators 31. The actuators 41 are coupled to each other at an
intermediate portion 4a at the substantially center between a first
end portion 4b and a second end portion 4c of the second actuator 4
in the longitudinal direction. As described above, two actuators 41
are coupled to each other at the intermediate portion 4a, and
therefore, drive force of two actuators 41 is bound so that the
second actuator 4 can exert great drive force. As described later,
the second actuator 4 is thermally expanded by heating by power
distribution to generate the drive force.
[0040] The first end portions 4b of two actuators 41 are coupled to
the first base member 21 at a bottom portion of the second recessed
portion 21b of the first base member 21. The second end portions 4c
of two actuators 41 are coupled to the second base member 22 at a
bottom portion of the fourth recessed portion 22b of the second
base member 22.
[0041] In a precise sense, the second actuator 4 does not extend
straight in the Y-direction, but slightly bends such that the
intermediate portion 4a protrudes rightward in the X-direction as a
drive direction of the second actuator 4 or slightly curves to
entirely expand rightward in the X-direction. Moreover, as
described above, the first actuator 3 and the second actuator 4
bend or curve with respect to the drive direction thereof, and
therefore, upon thermal expansion due to heating, do not bend or
curve to the opposite side of the drive direction. Thus, the first
actuator 3 and the second actuator 4 can reliably bend or curve in
the drive direction.
[0042] As described above, the first actuator 3 is arranged on the
left side in the X-direction in the shutter apparatus 1, the second
actuator 4 is arranged on the right side in the X-direction in the
shutter apparatus 1, and the first actuator 3 and the second
actuator 4 face each other as viewed in plane.
[0043] The drive target member 7 is arranged between the first
actuator 3 and the second actuator 4 facing each other. Moreover,
the first beam 5 and the second beam 6 are coupled to the drive
target member 7. The drive target member 7 is formed thinner than
other members forming the MEMS device 10. Thus, the mass of the
drive target member 7 is decreased, and a resonance frequency is
increased. Moreover, a metal film 71 such as an Au/Ti film is
formed across the entire surface of the drive target member 7. In
the shutter apparatus 1, the drive target member 7 functions as a
shutter configured to close or open a not-shown optical path. Thus,
the drive target member 7 is formed in a planar shape slightly
larger than the section of the optical path, and is specifically
formed in a circular shape. Moreover, a not-shown radiation portion
is formed at a coupling portion between each of the first beam 5
and the second beam 6 and the drive target member 7, and has the
function of releasing heat generated at the first actuator 3 upon
drive of the first actuator 3 and transmitted from the first
actuator 3 through the first beam 5 and/or heat generated at the
second actuator 4 upon drive of the second actuator 4 and
transmitted from the second actuator 4 through the second beam
6.
[0044] The first beam 5 is a rod-shaped member extending in the
X-direction. The first end portion 5a of the first beam 5 is
coupled to the intermediate portion 3a of the first actuator 3. The
second end portion 5b of the first beam 5 is coupled to the drive
target member 7.
[0045] The second beam 6 is a member having a turned-back
structure, and includes a first member 61 extending from the
intermediate portion 4a of the second actuator 4 to the vicinity of
the intermediate portion 3a of the first actuator 3 and a second
member 62 turned back to the second actuator 4 from an end portion
6a of the first member 61. The third end portion 6b (i.e., the base
end on a first member 61 side) of the second beam 6 is coupled to
the intermediate portion 4a of the second actuator 4. The fourth
end portion 6c (i.e., the tip end on a second member 62 side) of
the second beam 6 is coupled to the drive target member 7.
[0046] The second member 62 is a rod-shaped member extending in the
X-direction from the end portion 6a of the first member 61 and
having the substantially-same thickness as that of the first beam
5. The second member 62 is arranged in parallel with the first beam
5 on a side slightly lower than the first beam 5 in the
Y-direction. Note that "the first beam 5 and the second member 62
of the second beam are arranged in parallel" indicates, including
description below, that the first beam 5 and the second member 62
of the second beam are arranged with a substantially-parallel
relationship being maintained. Moreover, a portion 56 at which the
first beam 5 and the second member 62 of the second beam 6 are
arranged in parallel with each other will be sometimes referred to
as a "parallel arrangement portion 56." That is, the first beam 5
and the second member 62 of the second beam 6 are, in parallel,
coupled to the drive target member 7 from the same direction. In
other words, the tip end side of the first beam 5 is turned back at
the not-shown radiation portion, and therefore, at the coupling
portion between each of the first beam 5 and the second beam 6 and
the drive target member 7, the first beam 5 and the second member
62 of the second beam 6 are arranged in parallel.
[0047] The first member 61 is a member having, e.g., a hook shape
extending around the drive target member 7. Moreover, the first
member 61 has, at least at part thereof, a highly-elastic region to
have a higher elastic modulus than that of the second member 62,
and is formed wider, for example. As described later, even when the
second beam 6 is driven by the second actuator 4, the first member
61 remains in the hook shape with little elastic deformation, and
transmits the drive force of the second actuator 4 to the second
member 62. In addition, the highly-elastic region may be formed in
such a manner that part of the first member 61 is formed thicker
than the second member 62 or a metal film is formed on part of the
first member 61. Note that the second actuator 4 of the present
embodiment may be omitted, and the second beam 6 may be directly
coupled to the fixing portion 2. Specifically, the second beam 6
may include only the second member 62, and an end portion on a side
opposite to the fourth end portion 6c of the second member 62 may
be directly coupled to the fixing portion 2 (the first base member
21 or the second base member 22). In this case, the second member
62 may be directly coupled to the fixing portion 2 in a
substantially-straight state or a state with a slight curvature, or
the vicinity of the end portion on the side opposite to the fourth
end portion 6c of the second member 62 may be folded and coupled to
the fixing portion 2.
[0048] The first electrode 101 is a metal film formed on the upper
surface of the first base member 21, such as an Au/Ti film.
[0049] The second electrode 102 is a metal film formed on the upper
surface of the second base member 22, such as an Au/Ti film.
[0050] [Operation of Shutter Apparatus]
[0051] Subsequently, operation of the shutter apparatus 1
configured as described above will be described. FIG. 3 illustrates
a plan view of the shutter apparatus 1 in a drive state.
[0052] The shutter apparatus 1 is driven in such a manner that
voltage is applied to between the first electrode 101 and the
second electrode 102. When voltage is applied to between the first
electrode 101 and the second electrode 102, current flows in the
first actuator 3 and the second actuator 4 through the first base
member 21 and the second base member 22. At this point, Joule heat
is generated at the first actuator 3 and the second actuator 4 made
of the silicon material, and the first actuator 3 and the second
actuator 4 are instantaneously heated to 400 to 500.degree. C.
[0053] The first actuator 3 is thermally expanded by heating such
that the entire length thereof is extended. The positions of the
first end portion 3b and the second end portion 3c of the first
actuator 3 are fixed by the fixing portion 2, and therefore, the
intermediate portion 3a is pushed leftward in the X-direction as a
direction in which the intermediate portion 3a protrudes in
advance.
[0054] The second actuator 4 is also thermally expanded by heating
such that the entire length thereof is extended. The positions of
the first end portion 4b and the second end portion 4c of the
second actuator 4 are fixed by the fixing portion 2, and therefore,
the intermediate portion 4a is pushed rightward in the X-direction
as a direction in which the intermediate portion 4a protrudes in
advance.
[0055] When the intermediate portion 3a of the first actuator 3 is
pushed leftward in the X-direction, the first beam 5 coupled
thereto is entirely pulled leftward in the X-direction. Moreover,
when the intermediate portion 4a of the second actuator 4 is pushed
rightward in the X-direction, the second beam 6 coupled thereto is
entirely pulled rightward in the X-direction.
[0056] That is, the relative positions of the first end portion 5a
of the first beam 5 and the second end portion 6b of the second
beam 6 change in a direction in which these end portions are apart
from each other.
[0057] Even when the second beam 6 is entirely pulled rightward in
the X-direction, the first member 61 of the second beam 6 is little
elastically deformed, and therefore, most of pull force by the
second beam 6 is concentrated on the end portion 6a and changes to
the force of pushing the second member 62 rightward in the
X-direction. Of the first beam 5 and the second member 62 of the
second beam 6 arranged in parallel, the first beam 5 is, as a
result, pulled leftward in the X-direction, and the second member
62 of the second beam 6 is pushed rightward in the X-direction.
Thus, the second end portion 5b of the first beam 5 and the fourth
end portion 6c of the second beam 6 are driven diagonally to an
upper left side on an XY plane, the first beam 5 and the second
member 62 of the second beam 6 greatly curve or bend with different
curvatures, and the fourth end portion 6c of the second beam 6
pushes the drive target member 7. Meanwhile, the second end portion
5b of the first beam 5 pulls the drive target member 7. Thus, the
first member 61 of the second beam 6 slightly rotates
counterclockwise about the third end portion 6b on the XY plane,
and the drive target member 7 is pushed to a position on the XY
plane as illustrated in FIG. 5. Moreover, the second actuator 4 is
configured such that the multiple actuators are coupled to each
other as in the present embodiment, and therefore, as compared to a
case where the second actuator 4 includes one actuator,
counterclockwise rotational stiffness of the first member 61 of the
second beam 6 about the third end portion 6b on the XY plane can be
enhanced.
[0058] On the other hand, when voltage is no longer applied to
between the first electrode 101 and the second electrode 102,
current no longer flows in the first actuator 3 and the second
actuator 4, and the first actuator 3 and the second actuator 4 are
rapidly naturally cooled and the extended entire lengths thereof
return to original lengths. At this point, the intermediate portion
3a of the first actuator 3 pushed leftward in the X-direction is
pulled back rightward in the X-direction, and the intermediate
portion 4a of the second actuator 4 pushed rightward in the
X-direction is pulled back leftward in the X-direction.
[0059] When the intermediate portion 3a of the first actuator 3 is
pulled back rightward in the X-direction, the first beam 5 coupled
thereto is entirely pulled back rightward in the X-direction.
Moreover, when the intermediate portion 4a of the second actuator 4
is pulled back leftward in the X-direction, the second beam 6
coupled thereto is entirely pulled back leftward in the
X-direction.
[0060] That is, the relative positions of the first end portion 5a
of the first beam 5 and the second end portion 6b of the second
beam 6 change in a direction in which these end portions approach
each other.
[0061] Even when the second beam 6 is entirely pulled back leftward
in the X-direction, the first member 61 of the second beam 6 is
little elastically deformed, and therefore, most of drawing force
by the second beam 6 is concentrated on the end portion 6a and
changes to the force of pushing the second member 62 leftward in
the X-direction. Of the first beam 5 and the second member 62 of
the second beam 6 arranged in parallel, the first beam 5 is, as a
result, pushed rightward in the X-direction, and the second member
62 of the second beam 6 is pulled leftward in the X-direction.
Thus, the second end portion 5b of the first beam 5 and the fourth
end portion 6c of the second beam 6 are pushed back diagonally to a
lower right side on the XY plane, and the curved or bent first beam
5 and the curved or bent second member 62 of the second beam 6
return to original substantially-straight shapes. The first member
61 of the second beam 6 slightly rotates clockwise about the third
end portion 6b on the XY plane to return to an original position,
and the drive target member 7 returns to a position on the XY plane
as illustrated in FIG. 1.
[0062] Note that the first beam 5 may push the drive target member
7 while the second beam 6 pulls the drive target member 7.
[0063] As described above, by switching between voltage application
(the drive state of the shutter apparatus 1) to the first electrode
101 and the second electrode 102 and cancellation (a non-drive
state of the shutter apparatus 1) of such voltage application, the
position of the drive target member 7 on the XY plane is switched
as illustrated in FIGS. 5 and 1. The not-shown optical path is
arranged to overlap with the drive target member 7 illustrated in
FIG. 1 or the drive target member 7 illustrated in FIG. 5. The
position of the drive target member 7 is switched as illustrated in
FIGS. 5 and 1, and therefore, the drive target member 7 functions
as the shutter configured to close or open the not-shown optical
path. The not-shown optical path may be closed at such a position
that the drive target member 7 is not driven by the first actuator
3 and the second actuator 4, and may be opened at such a position
that the drive target member 7 is driven. Conversely, the not-shown
optical path may be opened at such a position that the drive target
member 7 is not driven by the first actuator 3 and the second
actuator 4, and may be closed at such a position that the drive
target member 7 is driven. Moreover, the shutter is a concept
including an optical attenuator configured to close or open part of
the optical path other than closing or opening of the optical
path.
[0064] [Findings Leading to the Invention of the Present
Application]
[0065] The inventor(s) of the present application et al. has found
that in a case where a high temperature operating life test
(hereinafter referred to as a "HTOL test") is performed for the
shutter apparatus 1 illustrated in FIGS. 1 and 2, the frequency of
defects greatly varies according to a product lot and a main defect
mode in this case is malfunction of the first and/or second
actuators 3, 4.
[0066] After further analysis had been conducted, it has been found
that in the shutter apparatus 1 determined as defective, the first
and/or second actuators 3, 4 are plastically deformed. Moreover, in
this case, it has been found that silicon oxide is precipitated
into the first silicon layer 210 at high density. This has assumed
that at the step of manufacturing the SOI substrate 200 or the step
of manufacturing the shutter apparatus 1, a silicon oxide
precipitate (hereinafter referred to as "precipitated oxide") is
generated in the first silicon layer 210 forming the first and/or
second actuators 3, 4 in thermal treatment of the SOI substrate
200. Moreover, in the case of the same number of silicon atoms, the
volume of the silicon oxide is greater than the volume of a silicon
single crystal. It has been assumed that dislocation occurs in the
first silicon layer 210 due to volume expansion upon generation of
the precipitated oxide and a dislocation density increases to a
certain value or more to cause plastic deformation of the first
and/or second actuators 3, 4.
[0067] Generally, it has been found that plastic deformation in a
solid crystal greatly depends on dislocation in the crystal,
particularly the density or size thereof. In a state in which the
first and/or second actuators 3, 4 deformed due to thermal
expansion are held for the long period of time, a state in which
stress is on the first silicon layer 210 forming the first and/or
second actuators 3, 4 is maintained. Moreover, in a state in which
heating and cooling of the first and/or second actuators 3, 4 are
often repeated, a change in the stress on the first silicon layer
210 forming the first and/or second actuators 3, 4 becomes great.
In any case, it has been assumed that when dislocation occurs at
high density in the first silicon layer 210, dislocation movement
in the crystal due to the stress is accelerated and plastic
deformation of the first and/or second actuators 3, 4 is
caused.
[0068] For this reason, the inventor(s) of the present application
et al. has first conducted study on whether or not a defect can be
reduced by reduction in an oxygen concentration in the first
silicon layer 210. If the oxygen concentration can be reduced, the
density of the precipitated oxide can be reduced, needless to say.
However, in terms of a manufacturing cost, the first silicon layer
is generally formed using a silicon substrate (hereinafter referred
to as a "CZ-silicon substrate") formed by a CZ method. It has been
known that in the CZ method, oxygen is taken into silicon melt
from, e.g., a quartz crucible and remains as an interstitial oxygen
element in a silicon single crystal. A general oxygen concentration
in the CZ-silicon substrate is about 5.times.10.sup.17/cm.sup.3 to
1.times.10.sup.18/cm.sup.3. For reducing the oxygen concentration
in the silicon substrate to a value equal to or less than this
value, such as equal to or less than 1.times.10.sup.17/cm.sup.3,
methods are conceivable, in which the silicon substrate is produced
by a floating zone method or a silicon single-crystal layer is
grown by an epitaxial growth method on the CZ-silicon substrate and
is used as the first silicon layer 210.
[0069] However, any method leads to a higher cost for manufacturing
the silicon substrate as compared to the CZ method, and it is
difficult to apply these methods to the MEMS device for which
inexpensive production is demanded.
[0070] For this reason, study has been conducted on whether or not
a dislocation occurrence density due to the precipitated oxide can
be reduced. It has been found that an interstitial oxygen atom and
a silicon atom contained in the silicon single crystal are
chemically bound to each other to generate the precipitated oxide
and the density, shape, and size of the precipitated oxide change
due to a temperature provided to the silicon single crystal. For
example, such behavior is described in detail in a Sueoka's first
research paper (Journal of Applied Physics, 1993, vol. 74, p.
5437-5444).
[0071] In the shutter apparatus 1 determined as defective, it is
assumed that the density of the precipitated oxide increases and
the size of the precipitated oxide increases in thermal treatment
of the SOI substrate 200 at the step of manufacturing the SOI
substrate 200 or the step of manufacturing the shutter apparatus 1
and a volume around the precipitated oxide greatly changes. It is
assumed that due to such a situation, high-density dislocation is
caused.
[0072] The above-described Sueoka's first research paper also
describes a finding regarding behavior of the interstitial oxygen
atom in the silicon single crystal, and the inventor(s) of the
present application et al. has focused on this point. According to
this finding, it has been found that the coefficients of diffusion
of the interstitial oxygen atom and the interstitial silicon atom
in the silicon single crystal are reversed at 1000.degree. C.
Considering this point, it is assumed that the diffusion flow rate
of the interstitial oxygen atom is higher than the diffusion flow
rate of the interstitial silicon atom at 1000.degree. C. or lower
and the diffusion flow rate of the interstitial silicon atom is
higher than the diffusion flow rate of the interstitial oxygen atom
at over 1000.degree. C. The diffusion flow rate described herein
indicates the amount of atoms passing through a unit area per unit
time, and a unit is represented by mol/cm.sup.2s, for example.
[0073] Based on this finding, the inventor(s) of the present
application et al. has found that a thermal treatment temperature
at the step of manufacturing the shutter apparatus 1 including the
displacement enlarging mechanism (the MEMS device) 10 is properly
controlled, and in this manner, the precipitated oxide and
occurrence of dislocation due to the precipitated oxide can be
reduced and failure due to reliability degradation of the
displacement enlarging mechanism (MEMS device) 10 and the shutter
apparatus 1 can be reduced. Details will be described below.
[0074] [SOI Substrate Manufacturing Method and Shutter Apparatus
Manufacturing Method]
[0075] FIGS. 4A to 4C illustrate sectional views at each step of a
SOI substrate manufacturing method according to the present
embodiment, and FIGS. 5A to 5D illustrate sectional views at each
step of a shutter apparatus manufacturing method.
[0076] First, the method for manufacturing the SOI substrate 200 as
a base material of the shutter apparatus 1 will be described. As
described above, in production of the MEMS device 10, a bonded SOI
substrate is generally used in terms of cost reduction.
[0077] As illustrated in FIG. 4A, a device substrate 110 and a base
substrate 130 are prepared. Both of the device substrate 110 and
the base substrate 130 are produced by the CZ method, and have a
predetermined concentration such as the above-described oxygen
concentration of about 5.times.10.sup.17/cm.sup.3 to
1.times.10.sup.18/cm.sup.3. An oxide film layer 120 is formed on a
surface of the device substrate 110. The oxide film layer 120 is
formed in such a manner that the device substrate 110 is thermally
oxidized (hereinafter referred to as "wet oxidation") in atmosphere
containing water vapor. The thickness of the oxide film layer 120
is equal to or greater than 1 .mu.m, and a thermal treatment
temperature in the wet oxidation is about 1050.degree. C. Further,
the device substrate 110 and the base substrate 130 are washed to
clean each surface.
[0078] Next, as illustrated in FIG. 4B, a principal surface of the
device substrate 110 and a principal surface of the base substrate
130 are bonded to each other at a room temperature and joined to
each other. Further, in this state, thermal treatment is performed
at a predetermined temperature. Joint between the device substrate
110 and the base substrate 130 becomes stronger by such thermal
treatment. The thermal treatment is performed in non-oxidizing
atmosphere, and the thermal treatment temperature is about
990.degree. C., for example.
[0079] As illustrated in FIG. 4C, the device substrate 110 bonded
to the base substrate 130 is divided in the thickness direction
such that a portion 140 of the device substrate 110 remains on the
base substrate 130. The portion 140 of the device substrate 110 is
polished, and is adjusted such that a silicon single-crystal layer
of the portion 140 has a desired thickness. The device substrate
110 is not necessarily divided in the thickness direction, and may
be polished and adjusted such that the silicon single-crystal layer
has the desired thickness.
[0080] In this manner, the SOI substrate 200 is obtained. The
silicon single-crystal layer of the portion 140 of the device
substrate 110 corresponds to the first silicon layer 210 as a
device layer, the oxide film layer 120 corresponds to the oxide
film layer 220 as a Box layer, and the base substrate 130
corresponds to the second silicon layer 230 as a handle layer.
[0081] Subsequently, the method for manufacturing the shutter
apparatus 1 will be described. Note that the sectional views of the
manufacturing steps illustrated in FIGS. 5A to 5E each correspond
to sectional views along the II-II line of FIG. 1.
[0082] The SOI substrate 200 produced by the method illustrated in
FIGS. 4A to 4C is prepared. For example, the thickness of the
device layer 210 is 30 .mu.m, the thickness of the Box layer 220 is
1 .mu.m, and the thickness of the handle layer 230 is 250 .mu.m. As
illustrated in FIG. 5A, the SOI substrate 200 is wet-oxidized, and
an oxide film layer 240 is formed on a surface of the first silicon
layer 210. A treatment temperature in thermal oxidization is higher
than 1000.degree. C., and thermal oxidization treatment is
performed at 1050.degree. C., for example. Moreover, thermal
oxidization is performed such that the thickness of the oxide film
layer 240 is about several tens to hundreds of nanometers. Note
that it may be enough that the oxide film layer 240 has a thickness
necessary as an etching mask in etching of the later-described
first silicon layer 210. Note that in description below, the step
illustrated in FIG. 5A, i.e., a step until the wet oxidization
after the SOI substrate 200 has been prepared, will be sometimes
referred to as a "substrate preparation step."
[0083] As illustrated in FIG. 5B, a not-shown resist pattern is
formed on a surface of the oxide film layer 240 by a
photolithography method, and is used as a mask to etch the oxide
film layer 240. In this manner, a mask pattern 241 is obtained.
[0084] Next, as illustrated in FIG. 5C, the mask pattern 241 is
used as an etching mask to etch the first silicon layer 210 as the
device layer. In this manner, an original form of the displacement
enlarging mechanism (the MEMS device) 10 including the fixing
portion 2, the first actuator 3, the second actuator 4, the first
beam 5, the second beam 6, and the drive target member 7 is
integrally formed in the first silicon layer 210. Note that in FIG.
5C, only part of the displacement enlarging mechanism (the MEMS
device) 10 is illustrated for the sake of convenience. Moreover,
after the displacement enlarging mechanism (the MEMS device) 10 has
been integrally formed, the mask pattern 241 is removed.
[0085] Particularly, the number of times of etching for the drive
target member 7 is greater than that for the other members by one,
and the drive target member 7 is formed thinly to have a thickness
of about 7 .mu.m. That is, although not shown in the figure, the
thickness of the drive target member 7 is less than the thicknesses
of the first actuator 3, the second actuator 4, the first beam 5,
and the second beam 6. Further, the first electrode 101 is formed
on a surface of the first base member 21, the second electrode 102
is formed on a surface of the second base member 22, and the metal
film 71 is formed on the surface of the drive target member 7. The
electrodes 101, 102 and the metal film 71 are, for example, Au/Ti
films including Ti with a thickness of 20 nm and Au with a
thickness of 300 nm.
[0086] As illustrated in FIG. 5D, when the original form of the
shutter apparatus 1 is formed in the first silicon layer 210 as the
device layer, a dummy wafer (not shown) is subsequently bonded to
the device layer 210 with wax (not shown), and a back layer, i.e.,
the Box layer 220 and the handle layer 230, of the shutter
apparatus 1 is etched. By such etching, the SOI substrate 200, in
this case a multilayer structure of the device layer 210, the Box
layer 220, and the handle layer 230, remains in the fixing portion
2, and only the first silicon layer 210 as the device layer remains
in the first actuator 3, the second actuator 4, the first beam 5,
the second beam 6, and the drive target member 7 as the movable
members in the displacement enlarging mechanism (MEMS device)
10.
[0087] Finally, the wax (not shown) and the dummy wafer (not shown)
are removed, and the shutter apparatus 1 is finalized.
[0088] [Relationship Among Thermal Treatment Temperature, Densities
of Precipitated Oxide Etc., and Defect Occurrence Frequency]
[0089] From the findings disclosed in the above-described Sueoka's
first research paper, the inventor(s) of the present application et
al. has derived the following hypothesis.
[0090] First, when the CZ-silicon substrate is thermally treated at
a temperature of equal to or lower than 1000.degree. C., the
interstitial oxygen atom in silicon is chemically bound to the near
silicon atom to generate a minute silicon oxide nucleus. Moreover,
in this temperature range, the interstitial oxygen atom is more
easily movable than the interstitial silicon atom. Thus, the
interstitial oxygen atom is thermally diffused, bound to the
above-described nucleus, and grown to the precipitated oxide, and
the size thereof increases. Moreover, due to growth of the
precipitated oxide, lattice strain therearound becomes greater. Due
to such lattice strain, dislocation occurs in the silicon single
crystal. Moreover, a dislocation density increases according to a
precipitated oxide occurrence density. As described above, plastic
deformation of the first and/or second actuators 3, 4 including the
first silicon layer 210 as the silicon single-crystal layer is
caused, and therefore, the failure due to reliability degradation
of the shutter apparatus 1 is caused.
[0091] On the other hand, when the CZ-silicon substrate is
thermally treated at a temperature higher than 1000.degree. C., the
precipitated oxide is generated as in the above-described case, but
in this temperature range, the interstitial silicon atom is more
easily movable than the interstitial oxygen atom. Thus, the
precipitated oxide is not grown much as compared to the
above-described case, and the occurrence density decreases. Thus,
it has been assumed that the dislocation occurrence density due to
the precipitated oxide can be suppressed low, plastic deformation
of the first and/or second actuators 3, 4 can be prevented, and the
failure due to reliability degradation of the shutter apparatus 1
can be reduced.
[0092] For this reason, the inventor(s) of the present application
et al. has focused on the thermal oxidization step illustrated in
FIG. 5A at the steps of manufacturing the shutter apparatus 1 as
illustrated in FIGS. 5A to 5D, and for different treatment
temperatures, has checked the precipitated oxide density and the
defect occurrence frequency of the shutter apparatus 1. As a
result, tendency showed that the defect occurrence frequency of the
shutter apparatus 1 is low in a case where a stacking fault density
is equal to or greater than 1.times.10.sup.4/cm.sup.2 and is much
lower in a case where the stacking fault density is equal to or
greater than 1.times.10.sup.4/cm.sup.2 and equal to or less than
5.times.10.sup.4/cm.sup.2. Moreover, tendency showed that the
defect occurrence frequency is low in a case where the precipitated
oxide density is equal to or less than 5.times.10.sup.5/cm.sup.2
and is much lower in a case where the precipitated oxide density is
equal to or less than 1.times.10.sup.5/cm.sup.2. Further, tendency
showed that the defect occurrence frequency is lower in a case
where both of the above-described conditions regarding the stacking
fault density and the precipitated oxide density are satisfied than
in a case where either one of the conditions is satisfied. These
conditions can be combined as necessary.
[0093] FIG. 6 illustrates one example of the precipitated oxide
density, the stacking fault density, and HTOL test results of the
shutter apparatus 1 in the case of different thermal treatment
temperatures at the thermal oxidization step illustrated in FIG.
5A. Note that in FIG. 6, observation of the precipitated oxide and
a stacking fault and derivation of the density were performed in a
state after the SOI substrate 200 has been thermally oxidized, and
the shutter apparatus 1 was not directly analyzed. Moreover, in
FIG. 6, the shape of an etch pit in the first silicon layer 210
formed by dipping in a Wright etching solution was observed, and
the number of etch pits corresponding to defects due to the
precipitated oxide was obtained within a field of view with a
predetermined area in an optical microscope and the precipitated
oxide density was derived. Moreover, actual defect type, shape, and
size were checked by the optical microscope or a scanning electron
microscope (SEM). Similarly, the presence or absence, density, and
size of the stacking fault were checked by the optical microscope
or the scanning electron microscope (SEM). Moreover, when the
percentage of the number of products determined as defective in the
HTOL test is equal to or lower than a predetermined value, it is
determined as successful (O.K.). When such a percentage exceeds the
predetermined value, it is determined as unsuccessful (N.G.).
[0094] As illustrated in FIG. 6, in a case where in a case where
the thermal treatment temperature is a low temperature of
1000.degree. C. (CONDITION A), the precipitated oxide density
exceeded 1.times.10.sup.6/cm.sup.2, and an evaluation result in the
HTOL test was N.G. Moreover, in a case where the thermal treatment
temperature is 1100.degree. C. exceeding 1000.degree. C. (CONDITION
B), the precipitated oxide density was equal to or less than
5.times.10.sup.5/cm.sup.2, and the evaluation result in the HTOL
test was O.K. In a case where the thermal treatment temperature is
1200.degree. C. exceeding 1000.degree. C. (CONDITIONS C to E), the
precipitated oxide density was less than 5.times.10.sup.4/cm.sup.2,
and the evaluation result in the HTOL test was O.K.
[0095] According to a Sueoka's second research paper (Koji Sueoka,
"Study on Oxide Precipitation in CZ-Si Single Crystal and
Oxidization-Induced Stacking Fault," Doctoral Degree Thesis of
Kyoto University, 1997, p. 2), the shape of precipitated oxide in a
silicon single crystal at a low temperature (650.degree. C. to
1050.degree. C.) is a plate shape, and the shape of precipitated
oxide in a silicon single crystal at a high temperature
(1000.degree. C. to 1250.degree. C.) is an isolated polyhedral
shape. In comparison between TEM bright-field images of these
precipitated oxides disclosed in the Sueoka's second research
paper, strain occurs around the plate-shaped precipitated oxide in
the silicon single crystal, whereas no such strain is conformed
around the isolated polyhedral precipitated oxide. From this
finding, it is assumed that the precipitated oxide generated by the
thermal treatment under Condition A has a higher density than those
under Conditions B to E and the shape thereof is also different and
is a plate shape. As described above, when the plate-shaped
precipitated oxide is generated, lattice strain therearound becomes
greater, and for this reason, dislocation easily occurs. On the
other hand, it is assumed that in the thermal treatment under
Conditions B to E, the isolated polyhedral precipitated oxide is
generated, but lattice strain therearound is small and the
dislocation density is smaller as compared to Condition A.
[0096] Moreover, the stacking fault is present in the first silicon
layer 210 separately from the precipitated oxide. Under Condition A
where the thermal treatment temperature is 1000.degree. C., no
stacking fault is confirmed within the predetermined field of view
by microscope observation. On the other hand, under Condition B
where the thermal treatment temperature is 1100.degree. C., the
stacking fault density is greater than 5.times.10.sup.4/cm.sup.2.
Under Conditions C to E where the thermal treatment temperature is
1200.degree. C., the stacking fault density is a value greater than
1.times.10.sup.4/cm.sup.2.
[0097] As generally known, the stacking fault is a fault caused due
to partial distortion of atom periodic arrangement on a crystal
lattice plane in a single crystal. On the other hand, dislocation
is a fault that displacement of atom arrangement of a crystal
lattice is in a linear shape. Under Condition A, it is assumed as
follows: the silicon atom present around the precipitated oxide and
the interstitial oxygen atom having moved by thermal diffusion are
bound to each other during thermal oxidization, and growth of the
precipitated oxide and occurrence of dislocation in the first
silicon layer 210 progress; whereas the interstitial silicon atom
is mainly consumed by occurrence and growth of the precipitated
oxide, and therefore, the stacking fault is not increased much.
[0098] Conditions B to E are at such a temperature that the
interstitial silicon atom is more easily movable than the
interstitial oxygen atom, and therefore, the interstitial silicon
atom is bound to the interstitial oxygen atom and moves around the
precipitated oxide without being taken into the precipitated oxide.
It is assumed that in the course of decreasing the temperature, the
interstitial silicon atom is arranged at an irregular position
between crystal lattice planes, and the number of times of
occurrence of the stacking fault increases as compared to the case
(Condition A) where the thermal treatment temperature is
1000.degree. C. Moreover, it has been known that the stacking fault
interacts with dislocation and suppresses dislocation movement in
the crystal. It is assumed that under Conditions B to E, the
evaluation results in the HTOL test are more favorable than that
under Condition A because of influence of the stacking fault
generated at a higher density than that under Condition A.
Summarizing the above-described results, it has been verified that
the above-described hypothesis is correct.
Advantages Effects Etc.
[0099] As described above, in the present embodiment, for
manufacturing the shutter apparatus 1 including the displacement
enlarging mechanism (the MEMS device) 10, the thermal oxidization
step of thermally oxidizing the SOI substrate 200 having the first
silicon layer 210 at the temperature higher than 1000.degree. C.,
i.e., at such a temperature that the diffusion flow rate of the
interstitial silicon atom in the silicon single crystal is higher
than the diffusion flow rate of the interstitial oxygen atom, and
the processing step of processing the SOI substrate 200 after the
thermal oxidization step are at least performed.
[0100] Focusing on the precipitated oxide density, for
manufacturing the shutter apparatus 1 including the displacement
enlarging mechanism (the MEMS device) 10, the substrate preparation
step of preparing the SOI substrate 200 having the first silicon
layer 210 whose precipitated oxide density is equal to or less than
5.times.10.sup.5/cm.sup.2 and the processing step of processing the
SOI substrate 200 after the substrate preparation step are at least
performed.
[0101] From another point of view, for manufacturing the shutter
apparatus 1 including the displacement enlarging mechanism (the
MEMS device) 10, the substrate preparation step of preparing the
SOI substrate 200 having the first silicon layer 210 whose stacking
fault density is equal to or greater than 1.times.10.sup.4/cm.sup.2
and the processing step of processing the SOI substrate 200 after
the substrate preparation step are at least performed. It can be
also said that the substrate preparation step of preparing the SOI
substrate 200 having the first silicon layer 210 whose precipitated
oxide density is equal to or less than 5.times.10.sup.5/cm.sup.2
and whose stacking fault density is equal to or greater than
1.times.10.sup.4/cm.sup.2 and the processing step of processing the
SOI substrate 200 after the substrate preparation step are at least
performed.
[0102] By thermal oxidization of the SOI substrate 200 at a
temperature higher than 1000.degree. C., generation of the
precipitated oxide in the first silicon layer 210 as the device
layer can be suppressed, and the density thereof can be reduced.
Thus, occurrence of dislocation due to the precipitated oxide can
be suppressed, and plastic deformation when the first and second
actuators 3, 4 as movable portions in the displacement enlarging
mechanism (the MEMS device) 10 are used for a long period of time
can be suppressed. As a result, operation reliability of the
displacement enlarging mechanism (the MEMS device) 10 and therefore
operation reliability of the shutter apparatus 1 can be enhanced.
Note that in terms of only precipitated oxide generation
suppression and occurrence density reduction, the thermal treatment
temperature may be around 1410.degree. C. as a silicon melting
point when exceeding 1000.degree. C., but is actually preferably
equal to or lower than 1270.degree. C. because substrate warpage
become greater or a new crystal fault is caused when exceeding
1300.degree. C.
[0103] From still another point of view, the SOI substrate 200
having the first silicon layer 210 satisfying a precipitated oxide
density of equal to or less than 5.times.10.sup.5/cm.sup.2, a
stacking fault density of 1.times.10.sup.4/cm.sup.2, or both is
prepared, and is processed to obtain the displacement enlarging
mechanism (the MEMS device) 10. Thus, dislocation occurrence and/or
dislocation movement due to the precipitated oxide can be
suppressed, and plastic deformation of the first and second
actuators 3, 4 as the movable portions in the displacement
enlarging mechanism (the MEMS device) 10 can be suppressed. As a
result, the operation reliability of the displacement enlarging
mechanism (the MEMS device) 10 and therefore the operation
reliability of the shutter apparatus 1 can be enhanced.
[0104] In recent years, in production of the SOI substrate 200 by a
bonding method, bonding strength is ensured while substrate warpage
and metal impurity diffusion are reduced, and therefore, tendency
shows that the thermal treatment temperature after bonding is
decreased. A temperature range selected at this point is often a
temperature around 1000.degree. C. and lower than 1000.degree.
C.
[0105] However, when the bonding temperature is lower than
1000.degree. C., the precipitated oxide density becomes higher in
the first silicon layer 210, and the dislocation occurrence density
due to the precipitated oxide also becomes higher, as described
above. Thus, by performing the thermal oxidization step or the
substrate preparation step described in the present embodiment,
even when the shutter apparatus 1 is manufactured using the SOI
substrate 200 produced by the final thermal treatment at a
temperature of equal to or lower than 1000.degree. C., plastic
deformation of the first and second actuators 3, 4 upon long-term
use thereof can be suppressed, and the operation reliability of the
displacement enlarging mechanism (the MEMS device) 10 and therefore
the operation reliability of the shutter apparatus 1 can be
enhanced.
[0106] After the wet oxidization step illustrated in FIG. 5A, in
other words, after the substrate preparation step, if the
precipitated oxide is generated at high density or dislocation due
to the precipitated oxide occurs at high density, plastic
deformation of the first and second actuators 3, 4 cannot be
suppressed, leading to degradation of the operation reliability of
the shutter apparatus 1. For this reason, it is necessary to avoid
the precipitated oxide from substantially growing at the step
performed after the wet oxidization step. Specifically, at the step
performed after the wet oxidization step (after the substrate
preparation step), the temperature applied to the SOI substrate 200
including the first silicon layer 210 needs to be set to a
temperature lower than 1000.degree. C., i.e., equal to or lower
than a temperature that the diffusion flow rate of the interstitial
oxygen atom in the silicon single crystal is higher than the
diffusion flow rate of the interstitial silicon atom and that the
precipitated oxide contained in the first silicon layer 210 does
not substantially grown. With reference to the above-described
Sueoka's first and second research papers, the temperature needs to
be set to equal to or lower than 600.degree. C.
[0107] Note that according to the findings disclosed in these
research papers, even when the temperature applied to the SOI
substrate 200 is 700.degree. C., it takes several tens to several
hundreds of hours to grow the precipitated oxide, and therefore,
treatment at equal to or lower than 700.degree. C. can be
incorporated into the step of manufacturing the shutter apparatus 1
while treatment time is taken into consideration.
[0108] The first silicon layer 210 as the device layer is formed
using the CZ-silicon substrate. Thus, the cost of the first silicon
layer 210 can be reduced, and the manufacturing cost of the
displacement enlarging mechanism (the MEMS device) 10 and therefore
the manufacturing cost of the shutter apparatus 1 can be
reduced.
[0109] In a case where the first silicon layer 210 is wet-oxidized
to form the oxide film layer 240, the above-described processing
step includes at least the mask pattern formation step of forming
the mask pattern 241 for processing the first silicon layer 210 and
the silicon layer processing step of patterning the first silicon
layer 210 by means of the mask pattern 241.
[0110] In the semiconductor micromachining technique, patterning of
a silicon oxide film is technically established, and is also a step
which can be performed at low cost. Since the oxide film layer 240
is used as described above, the mask pattern 241 having sufficient
etching resistance can be formed in subsequent etching of the first
silicon layer 210. This is because the thickness of the first
silicon layer 210 is about 30 .mu.m and the first silicon layer 210
cannot be processed into a desired shape due to a change in a mask
pattern shape or elimination of a pattern itself in etching of the
first silicon layer 210 in the case of a mask pattern including a
resist, as described above. Note that when the thickness of the
first silicon layer 210 is smaller than the above-described value,
such as about several the first silicon layer 210 may be patterned
using the mask pattern including the resist.
[0111] In some cases, instead of the above-described thermal
oxidization step for the SOI substrate 200, the thermal treatment
might be performed in non-oxidizing atmosphere or atmosphere
containing a slight amount of oxidized gas. For example, these
cases include a case where a high concentration of p-type or n-type
impurities is contained in the first silicon layer 210 and an
attempt is made to decrease the electric resistivity of the first
silicon layer 210 by activation of these impurities by the thermal
treatment. In this case, the thermal treatment is performed at the
temperature that the diffusion flow rate of the interstitial
silicon atom in the silicon single crystal is higher than the
diffusion flow rate of the interstitial oxygen atom. Thus, plastic
deformation of the first and second actuators 3, 4 can be also
suppressed, and the operation reliability of the displacement
enlarging mechanism (the MEMS device) 10 and therefore the
operation reliability of the shutter apparatus 1 can be also
enhanced.
[0112] The MEMS device 10 according to the present embodiment
includes at least the SOI substrate 200 having the first silicon
layer 210, the fixing portion 2 formed on the SOI substrate 200,
the first and second actuators 3, 4 as the thermal actuators
coupled to the fixing portion 2 and configured to generate heat by
current application to displace in the predetermined direction
according to the generated heat temperature, and the first and
second beams 5, 6 and the drive target member 7 coupled to the
first and second actuators 3, 4. Moreover, the condition where the
precipitated oxide density of the first silicon layer 210 is equal
to or less than 5.times.10.sup.5/cm.sup.2, the condition where the
stacking fault density of the first silicon layer 210 is equal to
or less than 1.times.10.sup.4/cm.sup.2, or both are satisfied.
[0113] Since the MEMS device 10 is configured as described above,
plastic deformation of the first and second actuators 3, 4 as the
thermal actuators including the first silicon layer 210 as a
component can be suppressed, and the operation reliability of the
displacement enlarging mechanism (the MEMS device) 10 can be
enhanced.
[0114] The shutter apparatus 1 according to the present embodiment
includes the displacement enlarging mechanism (the MEMS device) 10
having the above-described characteristics and the first and second
electrodes 101, 102 arranged on the fixing portion 2 and
electrically connected to the end portions 3b, 3c, 4b, 4c of the
first and second actuators 3, 4, and closes or opens the optical
path by the drive target member 7.
[0115] Since the shutter apparatus 1 is configured as described
above, the operation reliability of the displacement enlarging
mechanism (the MEMS device) 10 can be enhanced, and therefore, the
operation reliability of the shutter apparatus 1 itself can be
enhanced.
[0116] Note that as illustrated in FIG. 3, during operation of the
shutter apparatus 1, not only the first and second actuators 3, 4
but also first and second beams 5, 6 greatly curve, and stress is
on these components. Moreover, as described above, in drive of the
first and second actuators 3, 4, heat propagates to the first and
second beams 5, 6 from these actuators. Further, the first and
second beams 5, 6 also include the first silicon layer 210 as in
the first and second actuators 3, 4.
[0117] That is, by performing the manufacturing method of the
present embodiment described above, not only plastic deformation of
the first and second actuators 3, 4 but also plastic deformation of
the first and second beams 5, 6 can be suppressed. As a result, the
operation reliability of the displacement enlarging mechanism (the
MEMS device) 10 and therefore the operation reliability of the
shutter apparatus 1 can be enhanced. Moreover, in the displacement
enlarging mechanism (the MEMS device) 10 and the shutter apparatus
1, plastic deformation of the first and second beams 5, 6 can be
suppressed, and the operation reliability of the displacement
enlarging mechanism (the MEMS device) 10 and therefore the
operation reliability of the shutter apparatus 1 can be
enhanced.
[0118] Moreover, according to the displacement enlarging mechanism
(the MEMS device) 10, the first actuator 3 and the second actuator
4 are coupled to the fixing portion 2, the first end portion 5a of
the first beam 5 is coupled to the first actuator 3, the third end
portion 6b of the second beam 6 is coupled to the second actuator
4, and the drive target member 7 is coupled to the second end
portion 5b of the first beam 5 and the fourth end portion 6c of the
second beam 6. The first beam 5 pulls the drive target member 7
from the second end portion 5b in an extension direction of the
first beam 5 while the second beam 6 pushes the drive target member
7 from the fourth end portion 6c in an extension direction of the
second beam 6. Thus, the drive target member 7 coupled to these
beams 5, 6 is driven. That is, drive force of the first beam 5 and
the second beam 6 each driven by the first actuator 3 and the
second actuator 4 is combined to drive the drive target member 7.
Thus, the drive target member 7 can be greatly displaced by slight
displacement of the first actuator 3 and the second actuator 4 as
drive members.
[0119] In the shutter apparatus 1, when voltage is applied to
between the first electrode 101 and the second electrode 102,
current flows in the first actuator 3 and the second actuator 4,
and the first actuator 3 and the second actuator 4 are heated and
thermally deformed. Accordingly, the first beam 5 and the second
beam 6 are driven to drive the drive target member 7 coupled to
these two beams 5, 6. Thus, the drive target member 7 can be
greatly displaced by low-voltage application to between the first
electrode 101 and the second electrode 102.
Other Embodiments
[0120] The embodiment has been described above as an example of the
technique disclosed in the present application. However, the
technique disclosed in the present disclosure is not limited to
above, and is also applicable to embodiments to which changes,
replacements, additions, omissions, etc. are made as necessary.
Moreover, the components described above in the embodiment may be
combined to form a new embodiment. Further, the components
illustrated in the attached drawings and described in the detailed
description may include not only components essential for solving
the problems, but also components provided as an example of the
above-described technique and not essential for solving the
problems. For this reason, it shall not be acknowledged that
illustration and description of these non-essential components in
the attached drawings and the detailed description directly mean
that these non-essential components are essential.
[0121] Note that each of the first actuator 3 and the second
actuator 4 may include a single actuator. Moreover, the sizes and
structures of the first actuator 3 and the second actuator 4 are
not necessarily the same as each other, but may be different from
each other. The number of actuators forming the first actuator 3 or
the second actuator 4 may be different between the first actuator 3
and the second actuator 4, and for example, the first actuator 3
includes a single actuator and the second actuator 4 includes two
actuators. Moreover, the length of the member forming the first
actuator 3 and the length of the member forming the second actuator
4 may be different from each other. Further, only either one of the
first actuator 3 or the second actuator 4 may be driven to drive
the drive target member 7. In addition, either one of the first
actuator 3 or the second actuator 4 may be omitted. In this case,
only the beam coupled to the non-omitted actuator may have a
radiation structure.
[0122] In the above-described embodiment, the displacement
enlarging mechanism 10 included in the shutter apparatus 1 has been
described as one example of the MEMS device, but the MEMS device
according to the present invention is not limited to above as long
as the MEMS device includes the thermal actuator configured to
generate heat by Joule heat generated by current application to the
thermal actuator itself to displace in the predetermined direction
according to the generated heat temperature and the drive target
member coupled to the thermal actuator. Moreover, the
above-described "drive target member coupled to the thermal
actuator" includes not only the drive target member 7 illustrated
in FIGS. 1 to 3, but also the first and second beams 5, 6, needless
to say.
[0123] The MEMS device according to the present invention is also
applicable to apparatuses other than the shutter apparatus 1
described above in the embodiment. For example, the MEMS device may
be applied to an optical path changing apparatus configured to tilt
a mirror by a thermal actuator to change the direction of light
entering the mirror and a wavelength selection filter apparatus
configured to tilt a wavelength selection filter provided instead
of a mirror by a thermal actuator to modulate light having entered
an upper surface of the filter to emit light with a predetermined
wavelength from a lower surface of the filter.
INDUSTRIAL APPLICABILITY
[0124] According to the MEMS device manufacturing method according
to the present invention, occurrence of dislocation in the silicon
layer forming the MEMS device can be suppressed. Thus, the
operation reliability of the MEMS device can be enhanced, and the
MEMS device manufacturing method is particularly useful in
application to the MEMS device having the movable portion.
DESCRIPTION OF REFERENCE CHARACTERS
[0125] 1 Shutter Apparatus [0126] 2 Fixing Portion [0127] 10
Displacement Enlarging Mechanism (Mems Device) [0128] 21 First Base
Member [0129] 22 Second Base Member [0130] 23 Third Base Member
[0131] 3 First Actuator [0132] 3a Intermediate Portion [0133] 3b
First End Portion [0134] 3c Second End Portion [0135] 4 Second
Actuator [0136] 4a Intermediate Portion [0137] 4b First End Portion
[0138] 4c Second End Portion [0139] 5 First Beam [0140] 56 Parallel
Arrangement Portion [0141] 5a First End Portion [0142] 5b Second
End Portion [0143] 6 Second Beam [0144] 6b Third End Portion [0145]
6c Fourth End Portion [0146] 7 Drive Target Member [0147] 8
Coupling Member [0148] 101 First Electrode [0149] 102 Second
Electrode [0150] 200 SOI Substrate (Substrate) [0151] 210 First
Silicon Layer [0152] 220 Oxide Film Layer [0153] 230 Second Silicon
Layer
* * * * *